3310 - Liquid biopsy for mutational profiling of locoregional recurrent and/or metastatic squamous cell carcinoma of the head and neck

Background

Circulating tumor DNA (ctDNA) is a minimally-invasive technique to profile the tumor genome in different cancers but is still underexplored in squamous cell carcinoma of the head and neck (SCCHN). Our aim was to study the relevance of mutational profiling through liquid biopsy in SCCHN.

Methods

We collected plasma cell-free DNA from 39 patients (pts) with locoregional recurrent (LR) (n = 19) and/or metastatic (M) (n = 20) SCCHN. Targeted next-generation sequencing (Illumina, HiSeq4000) was performed on a custom panel of 604 genes (median coverage 1000x). When available (n = 18), a tumor biopsy was sequenced to evaluate the concordance of somatic mutations.

Results

M SCCHN pts had a higher probability of carrying ctDNA than LR pts (70% vs 31%, p = 0.0256). When ctDNA was detected, the quantity (estimated using the median of the allele frequencies (AF) of all the somatic plasma variants) did not differ between the two groups (median 2.3% vs 2.6%, p = 0.97). The concordance between liquid and solid biopsy showed that 26% of the plasma variants (14/54) were not detected in the matched tumors, reflecting some tumor heterogeneity. Conversely, 81% of the variants (168/208) identified in the solid tumors were not detected in the plasma. In a logistic regression model, the detection of the variant in the plasma was related to the metastatic status (Odds Ratio (OR) 4.50, p = 0.012), the variant AF in the tumor (OR = 5.31, p < 0.001) and the ctDNA quantity (OR = 4.13, p < 0.001). Classification of the variant as a driver or passenger based on TCGA and COSMIC was however not relevant (OR = 1.05, p = 0.948). As expected, TP53 was the most frequently mutated gene, both in liquid and solid biopsy. Of clinical interest, the plasma of 3 pts for whom no tumor was available carried 3 different potentially actionable PIK3CA mutations, including the hotspot E545K.

Conclusions

Detection of ctDNA using targeted NGS is feasible in pts with SCCHN, essentially for metastatic disease. Liquid biopsy alone can identify potentially actionable alterations but does not reflect the full landscape of the solid tumor. Presence of a solid tumor variant in the plasma is related to the variant AF, the metastatic status and the ctDNA quantity of the sample.

Clinical trial identification

NCT02139020.

Editorial acknowledgement

Legal entity responsible for the study

Cliniques Universitaires Saint-Luc - Université Catholique de Louvain, Brussels, Belgium.

Funding

Région Wallonne, Belgium - Fondation Louvain UCLouvain.

Disclosure

R. Galot: Advisory / Consultancy: Innate Pharma; Travel / Accommodation / Expenses: Astellas; Travel / Accommodation / Expenses: Amgen. J. Machiels: Honoraria (institution), Research grant / Funding (institution), Travel / Accommodation / Expenses: Pfizer; Advisory / Consultancy, Research grant / Funding (institution): Roche; Honoraria (institution), Research grant / Funding (institution): AstraZeneca; Advisory / Consultancy, Research grant / Funding (institution): Bayer; Advisory / Consultancy: Innate Pharma; Honoraria (institution), Research grant / Funding (institution): Merck Serono; Honoraria (institution), Advisory / Consultancy, Research grant / Funding (institution): Boehringer Ingelheim; Honoraria (institution), Research grant / Funding (institution), Travel / Accommodation / Expenses: BMS; Honoraria (institution), Research grant / Funding (institution): Novartis; Honoraria (institution), Research grant / Funding (institution): Janssen; Honoraria (institution), Research grant / Funding (institution): Incyte; Research grant / Funding (institution), Travel / Accommodation / Expenses: Amgen; Research grant / Funding (institution), Travel / Accommodation / Expenses, Uncompensated advisory role: MSD; Advisory / Consultancy, Data safety monitoring board with honoraria: Debio; Advisory / Consultancy, Data safety monitoring board with honoraria: Nanobiotix; Non-remunerated activity/ies, investigator and study coordinator: EORTC. All other authors have declared no conflicts of interest.